Search

Cloud computing has attracted scientists to deploy scientific applications by offering services such as Infrastructure-as-a-service (IaaS), Software-as-a-service (SaaS), and Platform-as-a-Service (PaaS). The research community is able to get access to resources on-demand within a short period of time. But, as the demand for cloud resources is dynamic in nature, this affects resource availability during scheduling. Hence, there is a need for efficient management of resources so that tasks can be scheduled based on their execution requirements. To provide a solution, a resource prediction based scheduling approach has been introduced in this paper which automates the resource allocation for scientific applications in a virtualized cloud environment. This research work focuses on the design of an optimized prediction based scheduling approach which maps the tasks of scientific application with the optimal VM by combining the features of swarm intelligence and TOPSIS. The proposed approach minimizes the execution time, cost, and SLA violation rate in comparison to existing scheduling heuristics. [ABSTRACT FROM AUTHOR]

Natural and environmental sciences are one of the scientific domains which seek a lot of attention as it requires high performance computation and large storage space. Cloud computing is such a platform that offers a customizable infrastructure where scientific applications can provision the required resources prior to execution. The elasticity characteristic of cloud computing and it's pay-as-you-go pricing model can reduce the resource usage cost for cloud client's. The various services offered by the cloud providers and the extravagant developments in the domain of cloud computing has attracted many scientists to deploy their applications on cloud. The change in number of tasks of scientific application directly affects the demand of cloud resources. Therefore, to handle the fluctuating demand of resources, there is a need to manage the resources in an efficient manner. This research work focuses on the design of a prediction based scheduling approach which maps the tasks of scientific application with the optimal VM by combining the features of swarm intelligence and multi-criteria decision making approach. The proposed approach improves the accuracy rate, minimizes the execution time, cost and service level agreement violation rate in comparison to existing scheduling heuristics. [ABSTRACT FROM AUTHOR]

Abstract Cloud Computing has become prime infrastructure for scientists to deploy scientific applications as it offers parallel and distributed environment for large-scale computations. During deployment, the significant prediction of resource usage is essential to achieve optimal scheduling for scientific applications. The existing resource prediction models fall short in providing reasonable accuracy because of high variances of cloud metrics. Therefore, to handle the varying cloud resource demands, it is necessary to accurately predict the future resource requirements for automatically provisioning the resources. In this paper, an Intelligent Regressive Ensemble Approach for Prediction (REAP) has been proposed which integrates feature selection and resource usage prediction techniques to achieve high performance. The effectiveness of proposed approach is evaluated in a real cloud environment by conducting a series of experiments. The experimental results show that the proposed approach outperforms the existing models by significantly improving the accuracy rate and reducing the execution time. The results are further validated by comparing the existing Learning Automata (LA) based ensemble approach with the proposed approach on the basis of error rate. Highlights • Regressive Ensemble Approach for Predicting (REAP) CPU usage of a scientific application is proposed. • Dataset is generated by simulating Cybershake on workflow management system. • A meta-heuristic, Genetic Algorithm is used for selecting the relevant features. • REAP improves the accuracy rate, reduces the execution time and error rate. [ABSTRACT FROM AUTHOR]

The state-of-the-art physics alliances have augmented various opportunities to solve complex real-world problems. These problems require both multi-disciplinary expertise as well as large-scale computational experiments. Therefore, the physics community needs a flexible platform which can handle computational challenges such as volume of data, platform heterogeneity, application complexity, etc. Cloud computing provides an incredible amount of resources for scientific users on-demand, thus, it has become a potential platform for executing scientific applications. To manage the resources of Cloud efficiently, it is required to explore the resource prediction and scheduling techniques for scientific applications which can be deployed on Cloud. This paper discusses an extensive analysis of scientific applications, resource predictions and scheduling techniques for Cloud computing environment. Further, the trend of resource prediction-based scheduling and the existing techniques have also been studied. This paper would be helpful for the readers to explore the significance of resource prediction-based scheduling techniques for physics-based scientific applications along with the associated challenges. [ABSTRACT FROM AUTHOR]

Cloud computing is the delivery of on-demand computing resources. Cloud computing has numerous applications in fields of education, social networking, and medicine. But the benefit of cloud for medical purposes is seamless, particularly because of the enormous data generated by the health care industry. This colossal data can be managed through big data analytics, and hidden patterns can be extracted using machine learning procedures. In particular, the latest issue in the medical domain is the prediction of heart diseases, which can be resolved through culmination of machine learning and cloud computing. Hence, an attempt has been made to propose an intelligent decision support model that can aid medical experts in predicting heart disease based on the historical data of patients. Various machine learning algorithms have been implemented on the heart disease dataset to predict accuracy for heart disease. Naïve Bayes has been selected as an effective model because it provides the highest accuracy of 86.42% followed by AdaBoost and boosted tree. Further, these 3 models are being ensembled, which has increased the overall accuracy to 87.91%. The experimental results have also been evaluated using 10,082 instances that clearly validate the maximum accuracy through ensembling and minimum execution time in cloud environment. [ABSTRACT FROM AUTHOR]